[Integrations](https://spiral.ai/integrations) › [Codex](https://spiral.ai/integrations/codex) › [ElevenLabs Voice Isolator](https://spiral.ai/integrations/codex/elevenlabs-voice-isolator)

# How to Use ElevenLabs Voice Isolator in Codex

ElevenLabs Voice Isolator extracts speech from recordings with background noise or music for podcast, interview, and film editing. It targets spoken dialogue; isolating singing from music is less reliable.

|  |  |
| --- | --- |
| Model | ElevenLabs Voice Isolator |
| Available through | ElevenLabs |
| Category | Audio |
| Price | $0.156 per minute. |
| Returns | audio |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with ElevenLabs Voice Isolator in Codex

- Use ElevenLabs Voice Isolator on assets/audio/interview.wav and save a separate cleaned track in assets/audio/cleaned/. Keep the original.
- Use ElevenLabs Voice Isolator to clean the podcast recordings in recordings/podcast/. Check that each is under one hour and 500 MB before processing.
- Extract a 30-second sample from recordings/demo-narration.wav where the background music overlaps speech, then process it with ElevenLabs Voice Isolator. Save both versions for comparison.
- Use ElevenLabs Voice Isolator on recordings/user-interview.mp3, then pass the cleaned audio to a separate transcription tool. Save the audio and transcript beside the original.
- Use ElevenLabs Voice Isolator to clean assets/voice/source.wav before our voice transformation step. Save the isolated recording separately for review.

## How to set up ElevenLabs Voice Isolator in Codex

Pick the way that suits you. Each takes about two minutes.

### Use Spiral's MCP server

1. **Add the hosted server**

```bash
codex mcp add spiral --url https://api.spiral.ai/mcp
```

2. **Sign in**: Codex opens the browser: sign in to Spiral and click Authorize. No API key. `codex mcp list` then shows Spiral.

```bash
codex mcp login spiral
```

### Use the Spiral skill

1. **Send Codex this message**: Your agent reads Spiral's skill and sets itself up.

```
set up https://spiral.ai/SKILL.md
```

2. **Sign in or paste a key**: Approve Spiral in the browser, or give the agent a key from the console.

### Use the Spiral CLI

1. **Install and sign in**: Installs the Spiral CLI, signs you in and adds its skill to your coding agents.

```bash
npx -y spiral-cli@latest init
```

### Call the Spiral API

1. **Create an API key**: In the console, under API keys.

2. **Run ElevenLabs Voice Isolator**: One request; the result comes back by polling or a webhook.

```bash
curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "elevenlabs",
    "model": "audio-isolation",
    "input": {
      "audio": "<audio>"
    }
  }'
```

Then paste one of the prompts above: Codex finds ElevenLabs Voice Isolator on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/codex

## ElevenLabs Voice Isolator tools and prices

- [ElevenLabs Voice Isolator](https://spiral.ai/tools/elevenlabs/audio-isolation) (ElevenLabs): $0.156 per minute.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| file_format | pcm_s16le_16, other | other | The format of input audio. Options are 'pcm_s16le_16' or 'other'. For `pcm_s16le_16`, the input audio must be 16-bit PCM at a 16kHz sample rate, single channel (mono), and little-endian byte order. |

## What can you do with ElevenLabs Voice Isolator in Codex?

Prepare cleaner dialogue assets for a project's podcasts, tutorials, or recorded demos. A coding agent can process project audio with ElevenLabs Voice Isolator and save separate outputs for review.

### Podcast dialogue cleanup

Use ElevenLabs Voice Isolator to reduce background noise in recorded podcast conversations before editing.

### Location interview cleanup

Extract speech from interviews recorded with unwanted ambient sound. Compare the isolated recording with the original before cutting quotes.

### Dialogue beneath background music

Isolate spoken dialogue from a mixed recording that contains music for a new edit.

### Meeting recording cleanup

Reduce background noise in recorded meetings to make spoken discussion easier to hear.

### Recorded stream cleanup

Clean the spoken audio from a recorded stream before preparing clips or an edited replay.

### Audio preparation for transcription

Isolate speech from a noisy recording, then send the result to a separate transcription tool.

### Source audio for voice transformation

Clean a noisy spoken recording before passing it to a separate voice changer.

## Overview

- [ElevenLabs](https://spiral.ai/providers/elevenlabs)
- [All Codex guides](https://spiral.ai/integrations/codex)

## ElevenLabs Voice Isolator in other agents

- [Amp](https://spiral.ai/integrations/amp/elevenlabs-voice-isolator)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/elevenlabs-voice-isolator)
- [Claude](https://spiral.ai/integrations/claude-desktop/elevenlabs-voice-isolator)
- [Claude Code](https://spiral.ai/integrations/claude-code/elevenlabs-voice-isolator)
- [Cline](https://spiral.ai/integrations/cline/elevenlabs-voice-isolator)
- [Cursor](https://spiral.ai/integrations/cursor/elevenlabs-voice-isolator)
- [Devin](https://spiral.ai/integrations/devin/elevenlabs-voice-isolator)
- [dots](https://spiral.ai/integrations/dots/elevenlabs-voice-isolator)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/elevenlabs-voice-isolator)
- [Gemini](https://spiral.ai/integrations/gemini/elevenlabs-voice-isolator)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/elevenlabs-voice-isolator)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/elevenlabs-voice-isolator)
- [goose](https://spiral.ai/integrations/goose/elevenlabs-voice-isolator)
- [Grok](https://spiral.ai/integrations/grok/elevenlabs-voice-isolator)
- [Grok Build](https://spiral.ai/integrations/grok-build/elevenlabs-voice-isolator)
- [Hermes](https://spiral.ai/integrations/hermes/elevenlabs-voice-isolator)
- [Instinct](https://spiral.ai/integrations/instinct/elevenlabs-voice-isolator)
- [Manus](https://spiral.ai/integrations/manus/elevenlabs-voice-isolator)
- [Muse](https://spiral.ai/integrations/muse/elevenlabs-voice-isolator)
- [Muse Code](https://spiral.ai/integrations/muse-code/elevenlabs-voice-isolator)
- [n8n](https://spiral.ai/integrations/n8n/elevenlabs-voice-isolator)
- [Notion](https://spiral.ai/integrations/notion/elevenlabs-voice-isolator)
- [OpenClaw](https://spiral.ai/integrations/openclaw/elevenlabs-voice-isolator)
- [opencode](https://spiral.ai/integrations/opencode/elevenlabs-voice-isolator)
- [Perplexity](https://spiral.ai/integrations/perplexity/elevenlabs-voice-isolator)
- [Raycast](https://spiral.ai/integrations/raycast/elevenlabs-voice-isolator)
- [VS Code](https://spiral.ai/integrations/vscode/elevenlabs-voice-isolator)
- [Windsurf](https://spiral.ai/integrations/windsurf/elevenlabs-voice-isolator)
- [Zapier](https://spiral.ai/integrations/zapier/elevenlabs-voice-isolator)

## More Codex guides

- [ElevenLabs Voice Changer Multilingual v2](https://spiral.ai/integrations/codex/elevenlabs-voice-changer-multilingual-v2)
- [ElevenLabs v4 Text to Speech](https://spiral.ai/integrations/codex/elevenlabs-v4-text-to-speech)
- [ElevenLabs Voice Changer](https://spiral.ai/integrations/codex/elevenlabs-voice-changer)
- [ElevenLabs v4 Turbo Text to Speech](https://spiral.ai/integrations/codex/elevenlabs-v4-turbo-text-to-speech)
- [ElevenLabs Turbo v2.5](https://spiral.ai/integrations/codex/elevenlabs-turbo-v2-5)
- [Elevenlabs Music v2.5](https://spiral.ai/integrations/codex/elevenlabs-music-v2-5)

## Questions about ElevenLabs Voice Isolator in Codex

**Can Codex use ElevenLabs Voice Isolator?**

Yes. Add Spiral to Codex as an MCP server, and Codex can find and run ElevenLabs Voice Isolator like any built-in tool.

**How much does ElevenLabs Voice Isolator cost in Codex?**

$0.156 per minute. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Codex plan is separate.

**Do I need an ElevenLabs account to use ElevenLabs Voice Isolator in Codex?**

No. Spiral runs ElevenLabs Voice Isolator on its own ElevenLabs account. You sign in to Spiral once and never paste a key.

**How long does it take to set up ElevenLabs Voice Isolator in Codex?**

About two minutes: add the hosted server, then sign in.

**Does Codex wait while ElevenLabs Voice Isolator runs?**

ElevenLabs Voice Isolator runs in the background. Codex starts the run, checks on it and gives you the audio when it's ready.

**What is ElevenLabs Voice Isolator best for?**

Cleaning spoken recordings before editing, especially podcasts, interviews, and film dialogue with unwanted background sound.

**Can it remove background music while keeping speech?**

Yes. Extracting spoken dialogue from recordings with background music is a supported use. Listen to the result before replacing your original track.

**Can it extract singing from a song?**

It may work, but ElevenLabs says Voice Isolator is not specifically optimized for extracting singing from music. Test a short passage before processing a full song.

**How long can a recording be?**

ElevenLabs documents a maximum length of one hour and a maximum file size of 500 MB. Keep each recording within both limits.

**What audio does it return?**

ElevenLabs documents MP3 as the default output. Keep the original recording alongside the isolated track for comparison.

